Schneier on Security
A blog covering security and security technology.
« A New Secure Hash Standard |
| Pipe Bombs Found, Disposed Of »
February 8, 2007
"Stop and Frisks" in New York
Interesting data from New York. The number of people stopped and searched has gone up fivefold since 2002, but the number of arrests due to these stops has only doubled. (The number of "summonses" has also gone up fivefold.)
Good data for the "Is it worth it?" question.
Posted on February 8, 2007 at 1:10 PM
• 30 Comments
To receive these entries once a month by e-mail, sign up for the Crypto-Gram Newsletter.
Unfortunately, this is a subscribers only article. has anyone else seen related articles?
Change the partner in the url from homepage to rssuserland, or some other valid partner.
There's a report just out that a police captain ordered his people to stop and frisk all black men at one subway station. That is, he ordered them to stop and frisk everyone matching the description of a robber who has struck several time - but the only consistent description is "black male".
> The number of people stopped [...] has gone up fivefold [...], but the number of arrests [...] has only doubled
Just curious - what would you deem a more acceptable or expected ratio? Say, this was more your field - computer security - and it turned out that scanning five times as many computers for naughty pictures turned up "just" two times as many. Especially if this increased search activity was public, wouldn't one expect some decrease of hits due to deterrence alone?
Certainly I have no disagreement with revulsion for random police searches and that sort of stuff. It's your quantitative sneer that makes me comment.
For those who don't have NYT registrations and are too lazy for bugmenot, added the story to the link here
According to the article, some of the increase in the number of searches may be better reporting. So the actual increase in the number of searches may be less than five times.
"Certainly I have no disagreement with revulsion for random police searches and that sort of stuff. It's your quantitative sneer that makes me comment."
When I sneer, I'm going to use a more forceful word than "only."
@ Frank Ch. Eigler
Your comparison with computer security is invalid. Computer security only deals with computers that you own or have permission to access. You already have implicit permission to search them, and the owner will have no objection. There is nearly no cost to these searches.
In the police case, the police are searching citizens. Many of these citizens are innocent. If you're going to search more people then you have to have a good justification because there is a high cost per search.
I am waiting for the Terrorists to gather in places such as this with fatal communicatable diseases for the sole purposes of targeting law enforcement.
What's not talked about it that fewer than 10% of the stops result in a summons or arrest. Is a 90% failure rate an acceptable tradeoff?
When Bruce Schneier sneers, entropy reverses itself.
I decided to concentrate on the numbers to see if that might help answer your question:
97,296 stopped and searched in 2002
508,540 stopped and searched in 2006 (but that might be partly due to more diligent counting)
2,819 arrested after stop and search in 2002
5,317 arrested after stop and search in 2006
1,461 summonses were issued after a stop and search in 2002
7,292 summonses were issued after a stop and search in 2006
There were 1,128 complaints about stop and search in 2003
There were 2,556 Complaints about stop and search in 2006
According to Paul J. Browne, the chief police spokesman, last years crime broke down as follows:
- 68.5% involved crimes in which the criminals were described as blacks
- 24.5% involved crimes in which the criminals were described as hispanic
- 5.3% involved crimes in which the criminals were described as white
- 1.7% unaccounted for in article
Also, last year:
- 55.2% of those stopped and searched were black
- 30.5% of those stopped and searched were hispanic
- 11.1% of those stopped and searched were white
- 3.2% unaccounted for in article
I come from the UK; I presume that summonses are issued for less serious crimes that don't require an on the spot arrest. Does the increase in summonses indicate more hassle from the police about trivial crime or have I misunderstood the situation?
What can I say about the "Is it worth it?" question - well not much.
I'm not the only one: "Councilman Peter F. Vallone Jr., chairman of the Council’s public safety committee, said his staff was unable to interpret the numbers immediately."
The only conclusions I can draw from this is that the NY police are doing more searches and catching more people with things that are not legal. To me, the "Is it worth it?" question comes down to a question of a trade-off: should the Fourth Amendment constitional protection be more strongly enforced or not?
Sorry, I really don't see how the information in the article helps here.
If 1% of the searches proved justified, then 99% proved unjustified.
Law enforcement can search anyone with probable cause, but 'probable' means 'more likely than not', not 'highly improbable'.
Since the police obviously did not have probable cause, then every one of those searches should be ruled illegal and the officers prosecuted -- and I do mean jailed.
Don't forget that some of those summonses and arrests may be for nothing worse than mouthing off to a police officer. In other words, they don't necessarily correspond to crimes caught or prevented.
Only in their warped little minds.
We are talking a lot about the costs (in civil liberties abrogated, police time spent on this instead of other more useful activities = opportunity costs, etc.) but not enumerating them. We can't actually do the cost/benefit without the cost...
SO roughly, 500k stop and search, assume one every half hour, 2 cops, $50/hr to have them on the street=$25M or $4700/arrest. Assume an additional cost of 4 hours to handle complaints: $0.5M. Of course, we can all argue about he cost metric for the civil liberties violation...
The more the government treats its citizens as suspects, the more citizens will learn to hate government.
Totalitarian regimes often have low crime committed by civilians, who are all afraid, but the crimes committed by the government increase dramatically. No matter what you think, the power of government to commit harm far exceeds the power of the entire citizenry.
And the "it's for your own good and safety" is the primary reason given by totalitarian regimes.
The terrorists are winning because the U.S. is becoming more like them, and is adding to fear and loathing, disrespect for authority, etc. Liberties lost are rarely recovered without a revolution, and revolutions against totalitarianism are truly bloody.
Freedom is a very fine thing. America should try it again.
I wonder if this "stop and frisk" is more of a "fishing expedition"; given the summonses issued, this is an income-producer for the city.
How long before we have someone doing this as a "loyalty test"-- if you whine and moan, you fail, for instance.
IOW, maybe this is an "are you a sheep yet?" test...
"revolutions against totalitarianism are truly bloody"
Nearly the entire Warsaw Pact stands as a shining counterexample to this claim. While I have no desire to lose freedoms in the first place, there is hope for getting them back without violence or bloodshed.
*Why* have random searches been increased? Is it truly a response to terrorism or is it for the financial gain of the police departments (and officers) through drug forfeiture laws?
The FBI wasn't able to get Al Capone for racketeering, so they contrived to get him for tax evasion.
Most of this really isn't about terrorism. It is an excuse to "deter criminal" activity.
Want to see it stop? Change the law so that, in the cases where anti-terrorism is the reason, only terrorism related issues can be charged. If they can't charge me for a brick of pot in my coat (illegal search and seizure), I'll bet you would see a rapid decline in "terrorist threats".
"Don't forget that some of those summonses and arrests may be for nothing worse than mouthing off to a police officer. In other words, they don't necessarily correspond to crimes caught or prevented."
I'd expect over 10% of people stopped for no reason to "mouth off", so if they're only arresting 1% of those they stop, the NY police are doing a pretty good job of keeping such incidents from escalating. Not that this says anything about whether that 1% includes any crimes caught or prevented besides being such an a**hole about an unjustified stop that even the most patient cop will bust you.
"*Why* have random searches been increased? Is it truly a response to terrorism or is it for the financial gain of the police departments (and officers) through drug forfeiture laws?"
If it is a response to terrorism, then it is stupid. How many active terrorists are there in the USA and what are the odds of catching them by stop and search based upon observed behaviour in a public place? I'd say the odds of success are near zero.
Clearly, policing in a totally reactive manner is not efficient, hence stop and searching people where there is a "probable cause" is (IMHO) justified. The problem is deciding what meets probable cause criteria and how to balance the rights of the individual against the requirement for protecting the public against criminals.
On the whole, my impression is that the police have become more assertive but it is very hard to say whether this reflects a required tightening up of street policing or something else. Without serious input from the police about policy, it appears to be very difficult to make an informed judgement.
This is a political issue which is why I questioned the real value of the article.
For those interested, here is more data, this from the UK (albeit from 2004).
Trends are similar so I am not sure deterrence is working. It did seem at a cursory glance (cursory!) that whites do not get treated any better than blacks or Asians though in terrorism-related stop-n-search, Asian numbers have gone up. Presumably because of an assumption that most are Muslims and a wrong one that all or most of them are Asians...
I wouldn't be suprised if at least 99% of arrests following the "stop and frisk" routine are for invented "crimes" which didn't harm anybody - like drug posession or talking back to a cop.
"revolutions against totalitarianism are truly bloody"
"Nearly the entire Warsaw Pact stands as a shining counterexample to this claim."
How much brutality, injustice, murder occured before the Warsaw Pact countries were allowed to break away? If it's murder or be murdered, well...
Are the London Police still searching for the right man to shoot five times in the head?
IF you are a young hispanic or black male you will be stopped and frisked. The act of walking down the street with a couple of friends I guess is enough "probable cause" but my question is where are these cops when the WHOLE neighborhood knows where the drug dealers are. you see them on the same blocks at the same time everyday carrying out illegal activities, Its not at all that you have to search for them. I guess its safer to stop and search a couple of young kids on their way home from school then it is to search the drug dealers. After speaking to many kids that have gone through this the one thing they all agree is that "you better keep your mouth shut if you want to go home". These questionable searches have become part of their daily lifes... this is truly sad
Schneier.com is a personal website. Opinions expressed are not necessarily those of BT.